HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

rocket-3B-GGUF

TheBloke/rocket-3B-GGUF

kind
gguf
quant
gguf (2-bit)
family
misc
modality
—
root model
pansophic/rocket-3B
provider
TheBloke
license
cc-by-sa-4.0
language
en
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
2,661
likes
40
created
—
last modified
2023-11-23T13:36:20+00:00 (2023)
huggingface
https://huggingface.co/TheBloke/rocket-3B-GGUF
tags
transformersggufstablelmenarxiv:2305.18290arxiv:2101.00027arxiv:2305.06161base_model:pansophic/rocket-3Bbase_model:quantized:pansophic/rocket-3Blicense:cc-by-sa-4.0region:us

Benchmark scores (8)

benchmarkcategoryscorecontextconfevidence
HellaSwagknowledge76.45HellaSwag (10-shot)medium| HellaSwag (10-shot) | 76.45 |
WinoGrandeknowledge67.8Winogrande (5-shot)medium| Winogrande (5-shot) | 67.8 |
TruthfulQAsafety54.38TruthfulQA (0-shot)medium| TruthfulQA (0-shot) | 54.38 |
AlpacaEvalchat45.71low| Falcon-Instruct 🦅 | 40B | SFT | 5.17 | 45.71 |
MMLUknowledge45.51MMLU (5-shot)medium| MMLU (5-shot) | 45.51 |
GSM8Kmath37.91GSM8K (5-shot)medium| GSM8K (5-shot) | 37.91 |
DROPreasoning24.49DROP (3-shot)medium| DROP (3-shot) | 24.49 |
MT-Benchchat2.75low| StableLM-Tuned-α 🦜 | 7B | SFT | 2.75 | - |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.